| { | |
| "best_metric": 0.4979843497276306, | |
| "best_model_checkpoint": "mikhail-panzo/zlm_b64_le5_s12000/checkpoint-1000", | |
| "epoch": 0.8375209380234506, | |
| "eval_steps": 500, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04187604690117253, | |
| "grad_norm": 4.788218975067139, | |
| "learning_rate": 2.4500000000000004e-07, | |
| "loss": 1.1819, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.08375209380234507, | |
| "grad_norm": 15.618935585021973, | |
| "learning_rate": 4.95e-07, | |
| "loss": 1.0042, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.12562814070351758, | |
| "grad_norm": 2.7821555137634277, | |
| "learning_rate": 7.4e-07, | |
| "loss": 0.9676, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.16750418760469013, | |
| "grad_norm": 3.759263038635254, | |
| "learning_rate": 9.85e-07, | |
| "loss": 0.9362, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.20938023450586266, | |
| "grad_norm": 3.906439781188965, | |
| "learning_rate": 1.235e-06, | |
| "loss": 0.8667, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.25125628140703515, | |
| "grad_norm": 4.397643566131592, | |
| "learning_rate": 1.485e-06, | |
| "loss": 0.8455, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.2931323283082077, | |
| "grad_norm": 2.94077467918396, | |
| "learning_rate": 1.7350000000000001e-06, | |
| "loss": 0.8376, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.33500837520938026, | |
| "grad_norm": 4.161074161529541, | |
| "learning_rate": 1.985e-06, | |
| "loss": 0.7521, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.3768844221105528, | |
| "grad_norm": 2.8088552951812744, | |
| "learning_rate": 2.235e-06, | |
| "loss": 0.7444, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.4187604690117253, | |
| "grad_norm": 2.62283992767334, | |
| "learning_rate": 2.4850000000000003e-06, | |
| "loss": 0.7129, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.4187604690117253, | |
| "eval_loss": 0.6066474914550781, | |
| "eval_runtime": 213.4911, | |
| "eval_samples_per_second": 39.763, | |
| "eval_steps_per_second": 4.974, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.46063651591289784, | |
| "grad_norm": 3.654665946960449, | |
| "learning_rate": 2.7350000000000005e-06, | |
| "loss": 0.6772, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.5025125628140703, | |
| "grad_norm": 5.1851887702941895, | |
| "learning_rate": 2.9850000000000002e-06, | |
| "loss": 0.6552, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.5443886097152428, | |
| "grad_norm": 4.017187595367432, | |
| "learning_rate": 3.2350000000000004e-06, | |
| "loss": 0.6506, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.5862646566164154, | |
| "grad_norm": 2.1161563396453857, | |
| "learning_rate": 3.485e-06, | |
| "loss": 0.6301, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.628140703517588, | |
| "grad_norm": 1.857313871383667, | |
| "learning_rate": 3.7350000000000002e-06, | |
| "loss": 0.6373, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.6700167504187605, | |
| "grad_norm": 3.6841330528259277, | |
| "learning_rate": 3.985000000000001e-06, | |
| "loss": 0.6305, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.711892797319933, | |
| "grad_norm": 2.6450912952423096, | |
| "learning_rate": 4.235e-06, | |
| "loss": 0.6169, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.7537688442211056, | |
| "grad_norm": 1.7241184711456299, | |
| "learning_rate": 4.485e-06, | |
| "loss": 0.5933, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.7956448911222781, | |
| "grad_norm": 2.602735996246338, | |
| "learning_rate": 4.735e-06, | |
| "loss": 0.583, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.8375209380234506, | |
| "grad_norm": 12.419028282165527, | |
| "learning_rate": 4.9850000000000006e-06, | |
| "loss": 0.5922, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.8375209380234506, | |
| "eval_loss": 0.4979843497276306, | |
| "eval_runtime": 214.0502, | |
| "eval_samples_per_second": 39.659, | |
| "eval_steps_per_second": 4.961, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 12000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 11, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9749197233971712.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |